Skip to content
体验新版
项目
组织
正在加载...
登录
切换导航
打开侧边栏
PaddlePaddle
DeepSpeech
提交
04bcb6a1
D
DeepSpeech
项目概览
PaddlePaddle
/
DeepSpeech
大约 1 年 前同步成功
通知
206
Star
8425
Fork
1598
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
Issue
245
列表
看板
标记
里程碑
合并请求
3
Wiki
0
Wiki
分析
仓库
DevOps
项目成员
Pages
D
DeepSpeech
项目概览
项目概览
详情
发布
仓库
仓库
文件
提交
分支
标签
贡献者
分支图
比较
Issue
245
Issue
245
列表
看板
标记
里程碑
合并请求
3
合并请求
3
Pages
分析
分析
仓库分析
DevOps
Wiki
0
Wiki
成员
成员
收起侧边栏
关闭侧边栏
动态
分支图
创建新Issue
提交
Issue看板
体验新版 GitCode,发现更多精彩内容 >>
提交
04bcb6a1
编写于
10月 28, 2021
作者:
小湉湉
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
fix rtf, fix inf input of speedyspeech, fix stft dir for 2.2.0
上级
6dbcd772
变更
4
隐藏空白更改
内联
并排
Showing
4 changed file
with
7 addition
and
5 deletion
+7
-5
parakeet/exps/gan_vocoder/multi_band_melgan/synthesize.py
parakeet/exps/gan_vocoder/multi_band_melgan/synthesize.py
+2
-1
parakeet/exps/gan_vocoder/parallelwave_gan/synthesize.py
parakeet/exps/gan_vocoder/parallelwave_gan/synthesize.py
+2
-1
parakeet/exps/speedyspeech/inference.py
parakeet/exps/speedyspeech/inference.py
+2
-2
parakeet/modules/stft_loss.py
parakeet/modules/stft_loss.py
+1
-1
未找到文件。
parakeet/exps/gan_vocoder/multi_band_melgan/synthesize.py
浏览文件 @
04bcb6a1
...
...
@@ -86,8 +86,9 @@ def main():
N
+=
wav
.
size
T
+=
t
.
elapse
speed
=
wav
.
size
/
t
.
elapse
rtf
=
config
.
fs
/
speed
print
(
f
"
{
utt_id
}
, mel:
{
mel
.
shape
}
, wave:
{
wav
.
shape
}
, time:
{
t
.
elapse
}
s, Hz:
{
speed
}
, RTF:
{
config
.
fs
/
speed
}
."
f
"
{
utt_id
}
, mel:
{
mel
.
shape
}
, wave:
{
wav
.
shape
}
, time:
{
t
.
elapse
}
s, Hz:
{
speed
}
, RTF:
{
rtf
}
."
)
sf
.
write
(
str
(
output_dir
/
(
utt_id
+
".wav"
)),
wav
,
samplerate
=
config
.
fs
)
print
(
f
"generation speed:
{
N
/
T
}
Hz, RTF:
{
config
.
fs
/
(
N
/
T
)
}
"
)
...
...
parakeet/exps/gan_vocoder/parallelwave_gan/synthesize.py
浏览文件 @
04bcb6a1
...
...
@@ -86,8 +86,9 @@ def main():
N
+=
wav
.
size
T
+=
t
.
elapse
speed
=
wav
.
size
/
t
.
elapse
rtf
=
config
.
fs
/
speed
print
(
f
"
{
utt_id
}
, mel:
{
mel
.
shape
}
, wave:
{
wav
.
shape
}
, time:
{
t
.
elapse
}
s, Hz:
{
speed
}
, RTF:
{
config
.
fs
/
speed
}
."
f
"
{
utt_id
}
, mel:
{
mel
.
shape
}
, wave:
{
wav
.
shape
}
, time:
{
t
.
elapse
}
s, Hz:
{
speed
}
, RTF:
{
rtf
}
."
)
sf
.
write
(
str
(
output_dir
/
(
utt_id
+
".wav"
)),
wav
,
samplerate
=
config
.
fs
)
print
(
f
"generation speed:
{
N
/
T
}
Hz, RTF:
{
config
.
fs
/
(
N
/
T
)
}
"
)
...
...
parakeet/exps/speedyspeech/inference.py
浏览文件 @
04bcb6a1
...
...
@@ -96,8 +96,8 @@ def main():
input_ids
=
frontend
.
get_input_ids
(
sentence
,
merge_sentences
=
True
,
get_tone_ids
=
True
)
phone_ids
=
input_ids
[
"phone_ids"
]
tone_ids
=
input_ids
[
"tone_ids"
]
phone_ids
=
input_ids
[
"phone_ids"
]
.
numpy
()
tone_ids
=
input_ids
[
"tone_ids"
]
.
numpy
()
phones
=
phone_ids
[
0
]
tones
=
tone_ids
[
0
]
...
...
parakeet/modules/stft_loss.py
浏览文件 @
04bcb6a1
...
...
@@ -51,7 +51,7 @@ def stft(x,
# calculate window
window
=
signal
.
get_window
(
window
,
win_length
,
fftbins
=
True
)
window
=
paddle
.
to_tensor
(
window
)
x_stft
=
paddle
.
tensor
.
signal
.
stft
(
x_stft
=
paddle
.
signal
.
stft
(
x
,
fft_size
,
hop_length
,
...
...
编辑
预览
Markdown
is supported
0%
请重试
或
添加新附件
.
添加附件
取消
You are about to add
0
people
to the discussion. Proceed with caution.
先完成此消息的编辑!
取消
想要评论请
注册
或
登录